Closed
Bug 1306594
Opened 8 years ago
Closed 8 years ago
test_peerConnection_audioRenegotiationInactiveAnswer.html is permafailing when tested properly
Categories
(Core :: WebRTC: Signaling, defect, P2)
Core
WebRTC: Signaling
Tracking
()
RESOLVED
FIXED
mozilla52
Tracking | Status | |
---|---|---|
firefox52 | --- | fixed |
backlog | webrtc/webaudio+ |
People
(Reporter: pehrsons, Assigned: pehrsons)
References
Details
Attachments
(1 file, 1 obsolete file)
There's a bug in AudioStreamHelper in our test framework so the checkAudioNotFlowing() path always passes. This is because we set up the WebAudio graph with an analyser and then check that the analyser outputs all zeroes immediately. Even before the graph has started receiving any audio data. When I wait a bit before starting to check the analyser I get a permafail. Here's a run on try to show this: https://treeherder.mozilla.org/#/jobs?repo=try&revision=d092941072ef I'll move the relevant patches to this bug.
Comment hidden (mozreview-request) |
Assignee | ||
Comment 2•8 years ago
|
||
Nils, do you want to take a look at the failure? Reproduces nicely everywhere - just import the patch.
Flags: needinfo?(drno)
Comment 3•8 years ago
|
||
Cool catch! The reason for the failure is pretty simple: Firefox continues to send RTP as you can see in the RTCP stats on about:webrtc on the wire.
Flags: needinfo?(drno)
Comment 4•8 years ago
|
||
Looking into why the offerer appears to ignore the a=inactive from the answer...
Assignee: nobody → drno
Comment 5•8 years ago
|
||
New bug 1306777 (to avoid confusing mozreview) contains the fix for this.
Assignee: drno → nobody
backlog: --- → webrtc/webaudio+
Rank: 25
Priority: -- → P2
Comment 6•8 years ago
|
||
mozreview-review |
Comment on attachment 8796520 [details] Bug 1306594 - Wait for the Analyser to start gathering data before analysing. https://reviewboard.mozilla.org/r/82356/#review81348
Attachment #8796520 -
Flags: review?(padenot) → review+
Comment 7•8 years ago
|
||
Just pushed bug 1306777 to autoland. Once it's merged to mc I think you should be able to rebase this and land it.
Comment hidden (mozreview-request) |
Comment 9•8 years ago
|
||
Obviously mozreview can't even handle the simple case of another person rebasing a changeset... (sigh). I'll kick off a try run and once that is green we can figure out who lands what.
Assignee | ||
Comment 10•8 years ago
|
||
I don't see anything obviously wrong in the try run. Landing.
Assignee | ||
Updated•8 years ago
|
Assignee: nobody → pehrson
Status: NEW → ASSIGNED
Assignee | ||
Updated•8 years ago
|
Attachment #8797357 -
Flags: review?(padenot)
Comment 11•8 years ago
|
||
Pushed by pehrsons@gmail.com: https://hg.mozilla.org/integration/autoland/rev/13feb326ae28 Wait for the Analyser to start gathering data before analysing. r=padenot
Assignee | ||
Updated•8 years ago
|
Attachment #8797357 -
Attachment is obsolete: true
Comment 12•8 years ago
|
||
bugherder |
https://hg.mozilla.org/mozilla-central/rev/13feb326ae28
Status: ASSIGNED → RESOLVED
Closed: 8 years ago
status-firefox52:
--- → fixed
Resolution: --- → FIXED
Target Milestone: --- → mozilla52
You need to log in
before you can comment on or make changes to this bug.
Description
•